Concerning about possible paper mill for genome-wide identification and characterization studies
Hi,
My main research area is in plant genetics (I'm a bit newer to the field) and I'm becoming pretty confused about the number of gene identification and characterization studies in plants.
For context, if you search up "gene identification and characterization" in pubmed or google scholar, you'll see tens of thousands of results that give the same types of article that pretty much do some combination of
gene identification via blast --> chromosomal localization --> multiple sequence alignment and phylogenetic trees --> cis-regulatory elements + protein-protein interaction graphs --> GO term analysis (which is already frequently done by the genome sequencing paper or some auto-annotating software)--> then gene expression profiling of X conditions (either they do it themselves or they retrieve some public screening data)
Maybe I'm misunderstanding this but isn't everything on this in-silico (except the expression profiling/stress condition test, which even that seems to be a "we need to do an easy, small wet-lab assay to pass the the reviewer's conditions") and couldn't it all be automated? I've heard of some tools like PlantTribes2, Spdev3.0 (or even random preprint pipelines like reactr and bat) but it's also possible for people to find/make their own Snakemake/Nextflow pipeline for this, which could automate large segments of this. I think those tools I mentioned are relatively newer, but seeing the vast volume of all the papers that have been going on for decades and also seeing that bioinformatics pipelines have existed for equally as much time, I feel like this is almost feels like an intentional (or maybe not, I don't know) paper mill operation.
Mostly seeing that these papers are coming from "X agriculture/forestry university" in some university in China but are still getting passed in peer-reviewed journals with decent impact factors (and they pretty much all cite each other as they're "building on" the methods framework).
Despite this technically being novel information (as one could simply mine out millions of papers for thousands and thousands of gene families in millions of cultivars and species) feels like me to be a violation of academia since it doesn't really feel creative, novel, or "research."
Thoughts on this?
EDIT: typos, examples, links